*Replication file: Divert when it does not hurt: The initiation of economic sanctions by US presidents from 1989 to 2015
*Hana Attia

clear
*replace line below with your working directory
cd "/Users/hanaattia/Dropbox/Mac (2)/Desktop/PaperI_23042021/RoIE/R&R/RoIE_R&R_2/Submission_Replication"
use data_20221023.dta

* you may need to install the following packages: acplot and sts13. Please run the code in the lines below
* ssc install acplot
* ssc install sts13


****************************************************Setup and descriptives****************************************************

destring quarter_year, replace
*help encode
encode quarter_year, generate (Quarteryear)
tset Quarteryear

generate lag_n_pres_imposition  = l.n_pres_imposition
generate lag_n_pres_not_intense = l.n_pres_not_intense
generate lag_n_pres_intense     = l.n_pres_intense

generate lag_quarterly_unemp_unadjusted = l.quarterly_unemp_unadjusted
generate lag_approval = l.approval
generate lag_interaction = lag_quarterly_unemp_unadjusted*pres_party_power_cong

*Table 1: Descriptives
sum n_pres_imposition n_pres_not_intense n_pres_intense quarterly_unemp_unadjusted pres_party_power_cong approval election_prox EU_imposition EU_intense

****************************************************Statistical analysis****************************************************

*Table 2: Poisson models estimating the count of sanctions
poisson n_pres_imposition lag_n_pres_imposition lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition
estat gof
poisson n_pres_not_intense lag_n_pres_not_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition
estat gof
poisson n_pres_intense lag_n_pres_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition
estat gof
poisson n_pres_intense lag_n_pres_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_intense
estat gof

*Table 3: Marginal effect of interaction term on the predicted count of non-costly sanctions
poisson n_pres_not_intense lag_n_pres_not_intense c.lag_quarterly_unemp_unadjusted##c.pres_party_power_cong lag_approval election_prox i.EU_imposition

margins, dydx(lag_quarterly_unemp_unadjusted) at (lag_n_pres_not_intense=1 pres_party_power_cong =  -21.5164  (mean) lag_approval (median) election_prox  EU_imposition=0) predict(xb)
margins, at (lag_n_pres_not_intense=1 lag_quarterly_unemp_unadjusted = 6.075701  pres_party_power_cong = -21.5164 (mean) lag_approval (median) election_prox EU_imposition = 0)
margins, at (lag_n_pres_not_intense=1 lag_quarterly_unemp_unadjusted = 7.637694  pres_party_power_cong = -21.5164 (mean) lag_approval (median) election_prox EU_imposition = 0)
display(1.113752 -  .6912992 )/ .6912992
*.6110998***

margins, dydx(lag_quarterly_unemp_unadjusted) at (lag_n_pres_not_intense=1 pres_party_power_cong =  -20.75147  (mean) lag_approval (median) election_prox  EU_imposition=0) predict(xb)
margins, at (lag_n_pres_not_intense=1 lag_quarterly_unemp_unadjusted = 6.075701  pres_party_power_cong = -20.75147 (mean) lag_approval (median) election_prox EU_imposition = 0)
margins, at (lag_n_pres_not_intense=1 lag_quarterly_unemp_unadjusted = 7.637694  pres_party_power_cong = -20.75147 (mean) lag_approval (median) election_prox EU_imposition = 0)
display(1.11902  -  .7058384)/ .7058384
*.58537705***

margins, dydx(lag_quarterly_unemp_unadjusted) at (lag_n_pres_not_intense=1 pres_party_power_cong =   -11.87175  (mean) lag_approval (median) election_prox  EU_imposition=0) predict(xb)
margins, at (lag_n_pres_not_intense=1 lag_quarterly_unemp_unadjusted = 6.063889  pres_party_power_cong =  -11.87175 (mean) lag_approval (median) election_prox EU_imposition = 0)
margins, at (lag_n_pres_not_intense=1 lag_quarterly_unemp_unadjusted = 7.637694  pres_party_power_cong =  -11.87175 (mean) lag_approval (median) election_prox EU_imposition = 0)
display(1.18203   - .8968867 )/.8968867
*.31792566**

margins, dydx(lag_quarterly_unemp_unadjusted) at (lag_n_pres_not_intense=1 pres_party_power_cong =   5.464836  (mean) lag_approval (median) election_prox  EU_imposition=0) predict(xb)
margins, at (lag_n_pres_not_intense=1 lag_quarterly_unemp_unadjusted = 6.075701  pres_party_power_cong =  5.464836 (mean) lag_approval (median) election_prox EU_imposition = 0)
margins, at (lag_n_pres_not_intense=1 lag_quarterly_unemp_unadjusted = 7.637694  pres_party_power_cong =  5.464836 (mean) lag_approval (median) election_prox EU_imposition = 0)
display(1.315457   -  1.440472)/ 1.440472
*-.08678753

margins, dydx(lag_quarterly_unemp_unadjusted) at (lag_n_pres_not_intense=1 pres_party_power_cong =   13.05067  (mean) lag_approval (median) election_prox  EU_imposition=0) predict(xb)
margins, at (lag_n_pres_not_intense=1 lag_quarterly_unemp_unadjusted = 6.075701  pres_party_power_cong =  13.05067 (mean) lag_approval (median) election_prox EU_imposition = 0)
margins, at (lag_n_pres_not_intense=1 lag_quarterly_unemp_unadjusted = 7.637694  pres_party_power_cong =  13.05067 (mean) lag_approval (median) election_prox EU_imposition = 0)
display(1.378481 -  1.770708  )/ 1.770708 
*-.22150857*

margins, dydx(lag_quarterly_unemp_unadjusted) at (lag_n_pres_not_intense=1 pres_party_power_cong =   17.92492  (mean) lag_approval (median) election_prox  EU_imposition=0) predict(xb)
margins, at (lag_n_pres_not_intense=1 lag_quarterly_unemp_unadjusted = 6.075701  pres_party_power_cong =  17.92492 (mean) lag_approval (median) election_prox EU_imposition = 0)
margins, at (lag_n_pres_not_intense=1 lag_quarterly_unemp_unadjusted = 7.637694  pres_party_power_cong =  17.92492 (mean) lag_approval (median) election_prox EU_imposition = 0)
display(1.420561 - 2.021838 )/2.021838
*-.29739128**

*Marginal effect EU sanctions discussed in the main analysis
margins, at (EU_imposition = 0 (mean) lag_n_pres_not_intense=1 lag_quarterly_unemp_unadjusted pres_party_power_cong lag_approval (median) election_prox)
margins, at (EU_imposition = 1 (mean) lag_n_pres_not_intense=1 lag_quarterly_unemp_unadjusted pres_party_power_cong lag_approval (median) election_prox)
display(1.834472 -   1.100237)/  1.100237 
*.66734258***


*Table 4: Marginal effect of interaction term on the predicted count of costly sanctions
poisson n_pres_intense lag_n_pres_intense c.lag_quarterly_unemp_unadjusted##c.pres_party_power_cong lag_approval election_prox i.EU_intense

margins, dydx(lag_quarterly_unemp_unadjusted) at (lag_n_pres_intense=0 pres_party_power_cong =  -21.5164  (mean) lag_approval (median) election_prox  EU_intense=0) predict(xb)
margins, at (lag_n_pres_intense=0 lag_quarterly_unemp_unadjusted = 6.075701  pres_party_power_cong = -21.5164 (mean) lag_approval (median) election_prox EU_intense = 0)
margins, at (lag_n_pres_intense=0 lag_quarterly_unemp_unadjusted = 7.637694  pres_party_power_cong = -21.5164 (mean) lag_approval (median) election_prox EU_intense = 0)
display( .5129532  - .3514676  )/ .3514676
* .45946084

margins, dydx(lag_quarterly_unemp_unadjusted) at (lag_n_pres_intense=0 pres_party_power_cong =  -20.75147  (mean) lag_approval (median) election_prox  EU_intense=0) predict(xb)
margins, at (lag_n_pres_intense=0 lag_quarterly_unemp_unadjusted = 6.075701  pres_party_power_cong = -20.75147 (mean) lag_approval (median) election_prox EU_intense = 0)
margins, at (lag_n_pres_intense=0 lag_quarterly_unemp_unadjusted = 7.637694  pres_party_power_cong = -20.75147 (mean) lag_approval (median) election_prox EU_intense = 0)
display( .5096155  -  .3552077 )/ .3552077
* .43469722

margins, dydx(lag_quarterly_unemp_unadjusted) at (lag_n_pres_intense=0 pres_party_power_cong =   -11.87175  (mean) lag_approval (median) election_prox  EU_intense=0) predict(xb)
margins, at (lag_n_pres_intense=0 lag_quarterly_unemp_unadjusted = 6.063889  pres_party_power_cong =  -11.87175 (mean) lag_approval (median) election_prox EU_intense = 0)
margins, at (lag_n_pres_intense=0 lag_quarterly_unemp_unadjusted = 7.637694  pres_party_power_cong =  -11.87175 (mean) lag_approval (median) election_prox EU_intense = 0)
display( .4724229  - .4011581 )/ .4011581
* .17764767

margins, dydx(lag_quarterly_unemp_unadjusted) at (lag_n_pres_intense=0 pres_party_power_cong =   5.464836  (mean) lag_approval (median) election_prox  EU_intense=0) predict(xb)
margins, at (lag_n_pres_intense=0 lag_quarterly_unemp_unadjusted = 6.075701  pres_party_power_cong =  5.464836 (mean) lag_approval (median) election_prox EU_intense = 0)
margins, at (lag_n_pres_intense=0 lag_quarterly_unemp_unadjusted = 7.637694  pres_party_power_cong =  5.464836 (mean) lag_approval (median) election_prox EU_intense = 0)
display( .4074504  - .5105516 )/ .5105516
* -.2019408

margins, dydx(lag_quarterly_unemp_unadjusted) at (lag_n_pres_intense=0 pres_party_power_cong =   13.05067  (mean) lag_approval (median) election_prox  EU_intense=0) predict(xb)
margins, at (lag_n_pres_intense=0 lag_quarterly_unemp_unadjusted = 6.075701  pres_party_power_cong =  13.05067 (mean) lag_approval (median) election_prox EU_intense = 0)
margins, at (lag_n_pres_intense=0 lag_quarterly_unemp_unadjusted = 7.637694  pres_party_power_cong =  13.05067 (mean) lag_approval (median) election_prox EU_intense = 0)
display( .381908  - .5670609 )/ .5670609
* -.32651325

margins, dydx(lag_quarterly_unemp_unadjusted) at (lag_n_pres_intense=0 pres_party_power_cong =   17.92492  (mean) lag_approval (median) election_prox  EU_intense=0) predict(xb)
margins, at (lag_n_pres_intense=0 lag_quarterly_unemp_unadjusted = 6.075701  pres_party_power_cong =  17.92492 (mean) lag_approval (median) election_prox EU_intense = 0)
margins, at (lag_n_pres_intense=0 lag_quarterly_unemp_unadjusted = 7.637694  pres_party_power_cong =  17.92492 (mean) lag_approval (median) election_prox EU_intense = 0)
display(  .3663472  - .6066294 )/.6066294 
* -.39609389

*Marginal effect EU intense sanctions
margins, at (EU_intense = 0 (mean) lag_n_pres_intense=0 lag_quarterly_unemp_unadjusted pres_party_power_cong lag_approval (median) election_prox)
margins, at (EU_intense = 1 (mean) lag_n_pres_intense=0 lag_quarterly_unemp_unadjusted pres_party_power_cong lag_approval (median) election_prox)
display(1.01897 -  .4451708)/ .4451708 
*1.2889417***


****************************************************Appendix****************************************************

*Table A2-1: Negative Binomial models
nbreg n_pres_imposition lag_n_pres_imposition lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition
nbreg n_pres_not_intense lag_n_pres_not_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition
nbreg n_pres_intense lag_n_pres_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition
nbreg n_pres_intense lag_n_pres_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_intense

*Figure A.3.1-1 & Table A3.1-1: Robustness checks allowing for autocorrelation
*all sanctions
poisson n_pres_imposition lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition
predict lpred 
gen pred=exp(lpred)
gen resid=(n_pres_imposition-pred)/sqrt(pred)
acplot resid, spike se lags(8) xlabel(0,1,2,3,4,5,6,7,8)

arpois n_pres_imposition lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox EU_imposition, ar(5) 
nlpred lpred2
gen pred2=exp(lpred2)
gen resid2=(n_pres_imposition-pred2)/sqrt(pred2)
acplot resid2, spike se lags(8) xlabel(0,1,2,3,4,5,6,7,8)

*non-costly
clear
*replace line below with your working directory
cd "/Users/hanaattia/Dropbox/Mac (2)/Desktop/PaperI_23042021/RoIE/R&R/RoIE_R&R_2/Submission_Replication"
use data_20221023.dta

destring quarter_year, replace
encode quarter_year, generate (Quarteryear)
tset Quarteryear

generate lag_quarterly_unemp_unadjusted = l.quarterly_unemp_unadjusted
generate lag_approval = l.approval
generate lag_interaction = lag_quarterly_unemp_unadjusted*pres_party_power_cong

poisson n_pres_not_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition
predict lpred 
gen pred=exp(lpred)
gen resid=(n_pres_not_intense-pred)/sqrt(pred)
acplot resid, spike se lags(8) xlabel(0,1,2,3,4,5,6,7,8)

arpois n_pres_not_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox EU_imposition, ar(6) 
nlpred lpred2
gen pred2=exp(lpred2)
gen resid2=(n_pres_not_intense-pred2)/sqrt(pred2)
acplot resid2, spike se lags(8) xlabel(0,1,2,3,4,5,6,7,8)

*costly
clear
*replace line below with your working directory
cd "/Users/hanaattia/Dropbox/Mac (2)/Desktop/PaperI_23042021/RoIE/R&R/RoIE_R&R_2/Submission_Replication"
use data_20221023.dta

destring quarter_year, replace
encode quarter_year, generate (Quarteryear)
tset Quarteryear

generate lag_quarterly_unemp_unadjusted = l.quarterly_unemp_unadjusted
generate lag_approval = l.approval
generate lag_interaction = lag_quarterly_unemp_unadjusted*pres_party_power_cong

poisson n_pres_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_intense
predict lpred 
gen pred=exp(lpred)
gen resid=(n_pres_intense-pred)/sqrt(pred)
acplot resid, spike se lags(8) xlabel(0,1,2,3,4,5,6,7,8)

arpois n_pres_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox EU_intense, ar(8) 
nlpred lpred2
gen pred2=exp(lpred2)
gen resid2=(n_pres_intense-pred2)/sqrt(pred2)
acplot resid2, spike se lags(8) xlabel(0,1,2,3,4,5,6,7,8)

*Table A3.2-1: Excluding sanctions against countries not prominent in the US public discourse (those below the 25th percentile)
clear
*replace line below with your working directory
cd "/Users/hanaattia/Dropbox/Mac (2)/Desktop/PaperI_23042021/RoIE/R&R/RoIE_R&R_2/Submission_Replication"
use data_20221023_NYT_25quant.dta

destring quarter_year, replace
encode quarter_year, generate (Quarteryear)
tset Quarteryear

generate lag_n_pres_imposition  = l.n_pres_imposition
generate lag_n_pres_not_intense = l.n_pres_not_intense
generate lag_n_pres_intense     = l.n_pres_intense

generate lag_quarterly_unemp_unadjusted = l.quarterly_unemp_unadjusted
generate lag_approval = l.approval
generate lag_interaction = lag_quarterly_unemp_unadjusted* pres_party_power_cong

poisson n_pres_imposition lag_n_pres_imposition lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition
estat gof
poisson n_pres_not_intense lag_n_pres_not_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition
estat gof
poisson n_pres_intense lag_n_pres_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_intense
estat gof

nbreg n_pres_imposition lag_n_pres_imposition lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition
nbreg n_pres_not_intense lag_n_pres_not_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition
nbreg n_pres_intense lag_n_pres_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_intense


*Table A3.2-2: Excluding sanctions against countries not prominent in the US public discourse (those below the 50th percentile)
clear
*replace line below with your working directory
cd "/Users/hanaattia/Dropbox/Mac (2)/Desktop/PaperI_23042021/RoIE/R&R/RoIE_R&R_2/Submission_Replication"
use data_20221023_NYT_50quant.dta

destring quarter_year, replace
encode quarter_year, generate (Quarteryear)
tset Quarteryear

generate lag_n_pres_imposition  = l.n_pres_imposition
generate lag_n_pres_not_intense = l.n_pres_not_intense
generate lag_n_pres_intense     = l.n_pres_intense

generate lag_quarterly_unemp_unadjusted = l.quarterly_unemp_unadjusted
generate lag_approval = l.approval
generate lag_interaction = lag_quarterly_unemp_unadjusted* pres_party_power_cong

set more off
poisson n_pres_imposition lag_n_pres_imposition lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition
estat gof
poisson n_pres_not_intense lag_n_pres_not_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition
estat gof
poisson n_pres_intense lag_n_pres_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_intense
estat gof

*Table A3.3-1: Controlling for partisanship
clear
*replace line below with your working directory
cd "/Users/hanaattia/Dropbox/Mac (2)/Desktop/PaperI_23042021/RoIE/R&R/RoIE_R&R_2/Submission_Replication"
use data_20221023.dta

destring quarter_year, replace
encode quarter_year, generate (Quarteryear)
tset Quarteryear

generate lag_n_pres_imposition  = l.n_pres_imposition
generate lag_n_pres_not_intense = l.n_pres_not_intense
generate lag_n_pres_intense     = l.n_pres_intense

generate lag_quarterly_unemp_unadjusted = l.quarterly_unemp_unadjusted
generate lag_approval = l.approval
generate lag_interaction = lag_quarterly_unemp_unadjusted* pres_party_power_cong

set more off
poisson n_pres_imposition lag_n_pres_imposition lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition democrat
estat gof
poisson n_pres_not_intense lag_n_pres_not_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition democrat
estat gof
poisson n_pres_intense lag_n_pres_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_intense democrat
estat gof

*Table A3.4-1: Robust clustered standard errors on US administration level
gen group = .
replace group = 1 if pres2==1
replace group = 2 if pres3==1
replace group = 3 if pres4==1
replace group = 4 if pres5==1

poisson n_pres_imposition lag_n_pres_imposition lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition, vce(cluster group)
estat gof
poisson n_pres_not_intense lag_n_pres_not_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_imposition, vce(cluster group)
estat gof
poisson n_pres_intense lag_n_pres_intense lag_quarterly_unemp_unadjusted pres_party_power_cong lag_interaction lag_approval election_prox i.EU_intense, vce(cluster group)
estat gof

*Table A3.5-1: Logit model
clear
*replace line below with your working directory
cd "/Users/hanaattia/Dropbox/Mac (2)/Desktop/PaperI_23042021/RoIE/R&R/RoIE_R&R_2/Submission_Replication"
use dyadic_analysis_20221106.dta
set more off
logit imposition c.annual_unemp_unadjusted_lagged##c.pres_party_power_cong approval_lagged presidential_elections logged_annual_trade_lagged logged_eco_aid_lagged logged_mil_aid_lagged v2x_libdem_lagged log_NYT_yearly_t_1 t1_imposition t2_imposition t3_imposition, vce(cluster ccode)
logit imposition_not_intense c.annual_unemp_unadjusted_lagged##c.pres_party_power_cong approval_lagged presidential_elections logged_annual_trade_lagged logged_eco_aid_lagged logged_mil_aid_lagged v2x_libdem_lagged log_NYT_yearly_t_1 t1_imposition t2_imposition t3_imposition, vce(cluster ccode)
logit imposition_intense c.annual_unemp_unadjusted_lagged##c.pres_party_power_cong approval_lagged presidential_elections logged_annual_trade_lagged logged_eco_aid_lagged logged_mil_aid_lagged v2x_libdem_lagged log_NYT_yearly_t_1 t1_imposition t2_imposition t3_imposition, vce(cluster ccode)

*Table A3.5-2: Logit model clustering SE on the presidential administration level
set more off
logit imposition c.annual_unemp_unadjusted_lagged##c.pres_party_power_cong approval_lagged presidential_elections logged_annual_trade_lagged logged_eco_aid_lagged logged_mil_aid_lagged v2x_libdem_lagged log_NYT_yearly_t_1 t1_imposition t2_imposition t3_imposition, vce(cluster pres)
logit imposition_not_intense c.annual_unemp_unadjusted_lagged##c.pres_party_power_cong approval_lagged presidential_elections logged_annual_trade_lagged logged_eco_aid_lagged logged_mil_aid_lagged v2x_libdem_lagged log_NYT_yearly_t_1 t1_imposition t2_imposition t3_imposition, vce(cluster pres)
logit imposition_intense c.annual_unemp_unadjusted_lagged##c.pres_party_power_cong approval_lagged presidential_elections logged_annual_trade_lagged logged_eco_aid_lagged logged_mil_aid_lagged v2x_libdem_lagged log_NYT_yearly_t_1 t1_imposition t2_imposition t3_imposition, vce(cluster pres)

*Table A3.5-3: Logit model controlling for EU sanctions
set more off
logit imposition c.annual_unemp_unadjusted_lagged##c.pres_party_power_cong approval_lagged presidential_elections logged_annual_trade_lagged logged_eco_aid_lagged logged_mil_aid_lagged v2x_libdem_lagged log_NYT_yearly_t_1 EU_imposition t1_imposition t2_imposition t3_imposition, vce(cluster ccode)
logit imposition_not_intense c.annual_unemp_unadjusted_lagged##c.pres_party_power_cong approval_lagged presidential_elections logged_annual_trade_lagged logged_eco_aid_lagged logged_mil_aid_lagged v2x_libdem_lagged log_NYT_yearly_t_1 EU_imposition t1_imposition t2_imposition t3_imposition, vce(cluster ccode)
logit imposition_intense c.annual_unemp_unadjusted_lagged##c.pres_party_power_cong approval_lagged presidential_elections logged_annual_trade_lagged logged_eco_aid_lagged logged_mil_aid_lagged v2x_libdem_lagged log_NYT_yearly_t_1 EU_imposition t1_imposition t2_imposition t3_imposition, vce(cluster ccode)
